Anastasios Lamproudis

 Stockholm, Sweden @- Anastasios Lamproudis  github.com/TLampr Anastasios Lamproudis Education Kungliga Tekniska Högskolan Stockholm, Sweden M.Sc. in Machine Learning - – Thesis: “Word embeddings for the classification of main conditions with samples of patient text describing symptoms” Aristotle University of Thessaloniki Thessaloniki, Greece B.Sc. in Physics - Experience Doktor24 02/2020–10/2020 Machine Learning master thesis Internship Stockholm, Sweden – Title: Word embeddings for the classification of main conditions with samples of patient text describing symptoms – Tools: Pytorch Scikit-learn Improved the application’s performance by extracting, evaluating, and comparing text representations from Language models including Fasttext, ELMO, BERT, DistilBERT, ALBERT, XLM, and XLM-R. In a second stage further explored the zero-shot properties of multilingual Language models. Machine Learning Deep Learning Word embeddings Fasttext Text classification Python Transfer Learning ELMO BERT Zero-shot Learning DistilBERT ALBERT NLP XLM Language Models XLM-R SQL Projects Siamese and 3D Convolutional Neural Networks for Text-independent Speaker Verification Tools: Pytorch Scikit-learn Google Cloud – Applied Deep Learning techniques such as 3D Convolutional Neural networks and Siamese networks to create a model to perform Speaker verification independently of the spoken phrase. Machine Learning Deep Learning Speech Recognition Speaker verification Python CNN Audio Processing Siamese Networks NeurIPS reproducibility challenge - 2019 / De-noising Network: Toward Blind Noise Modeling and Removal Tools: Tensorflow 2.0 Scikit-learn Google cloud – Application of Deep Learning, optimisation, and variational techniques to perform Image De-noising. As part of the NeurlPS reproducibility challenge for the paper “Variational De-noising Network: Toward Blind Noise Modeling and Removal” Machine Learning Deep Learning Computer Vision Image de-noising Image processing Variational Inference Python U-Net CNN Page 1 of 2 Stability prediction of mRNA as part of Stanford University’s Kaggle competition for the covid-19 vaccine / “OpenVaccine: COVID-19 mRNA Vaccine Degradation Prediction” Tools: Pytorch Pandas – International competition as part of the research for the covid-19 vaccine. Applied state of the art Deep Learning approaches such as transformers and self-attention to design a network that can predict likely degradation rates at each base of an RNA molecule. Machine Learning Deep Learning Regression Time-series Sequential samples feature selection Python self-attention transformers LSTMs Covid-19 Image inpainting for irregular holes using partial convolutions by reproducing the homonymous paper published in 2018. Tools: Pytorch OpenCV Google cloud PIL/Pillow – Applied Deep Learning techniques such as U-Nets and partial convolutions to design a model that can repair randomly destroyed image segments. Performed as part of the reproduction of the paper “Image Inpainting for Irregular Holes using Partial Convolutions”. Machine Learning Deep Learning Computer Vision Image reconstruction Image processing Python U-Net CNN ResNet Hierarchical image classification with self-attention Tools: Tensorflow 2.0 Google cloud – Applied Deep Learning techniques such as self-attention, ResNet, and Convolutional Neural networks to a hierarchical classification task. Machine Learning Deep Learning Image processing Python Transfer Learning self-attention CNN Computer Vision Image classification ResNet Course Work Machine Learning Physics As part of the M.Sc. studies: As part of the B.Sc studies: Probability Theory, Deep Learning, Artificial Intelligence, Machine Learning, Data Science, Artificial Neural Networks, Speech & Speaker recognition, Natural Language Processing, Computer Vision Linear Algebra, Calculus, Mechanics, Quantum Mechanics, Astronomy, Electromagnetism, Thermodynamics Skills Languages • Programming Languages: • Greek: Native speaker Confident: Python Familiar: Matlab Beginner: Java, Scala, C++ • Tools and Frameworks: Confident: Pytorch, Sci-kit Learn, Numpy, Pandas Familiar: Tensorflow 2.0 Beginner: Apache Spark • English: Fluent, C1 proficiency level • German: Basic, B2 proficiency level • Italian: Basic, B2 proficiency level • Databases: Beginner: SQL Volunteer Work • Volunteer Tutor Autumn 2017–Spring 2018 Volunteer tutor for high school students in Physics, Algebra, Chemistry, and Biology. Page 2 of 2